fix: torch_float should return float, not int

[LincolnBurrows2017]

Description

The torch_float function in src/transformers/utils/generic.py was incorrectly returning int(x) in two places where it should return float(x):

1. When torch is not available (fallback case)
2. When not in a tracing context with a torch tensor

This is inconsistent with the function's name and purpose (converting to float32). The similar function torch_int correctly returns int(x).

Fix

Changed int(x) to float(x) in both locations.

---

Bug originally identified during AI contribution task.

[github-actions]

[For maintainers] Suggested jobs to run (before merge)

run-slow: doge

[xenova]

thanks! I'm not sure about the other changes in this PR, but the fix for int() when it should be float() is correct. Surprised it went unnoticed for so long.